[petsc-users] GAMG

Mark Adams mfadams at lbl.gov
Wed Jun 10 11:28:14 CDT 2015


Yes, lets get this back on the list.

On Wed, Jun 10, 2015 at 12:01 PM, Young, Matthew, Adam <may at bu.edu> wrote:

>  Ah, oops - I was looking at the v 3.5 manual. I am certainly interested
> in algorithmic details if there are relevant papers. My main interest right
> now is determining if this method is appropriate for my problem.
>

Jed mentioned that this will not work well out of the box, as I recall.  It
looks like very high anisotropy.  There are heuristics in GAMG to deal with
anisotropy, but I would not trust them on unstructured grids.  You would
want to play with '-pc_gamg_threshold x', x in [0 - 0.1].

My advisor used SuperLU on a sequential version of this problem years ago,
> but SuperLU_dist appears to not scale well.
>
> yep,


>  Before any of that, though: Would you rather I redirect this to the
> petsc-users list? I emailed you directly because I worried that the
> discussion of my problem physics was detracting from PETSc-specific
> conversation but maybe this is a conversation better had with the entire
> PETSc group.
>
>
You might look in the appendix of the "Multigrid" book by Trottenberg, et
al.  They have a listing of lots of physics problems and way to deal with
them, with multigrid.  Maybe they have something like your problem.

Mark


>  --Matt
>
>  --------------------------------------------------------------
> Matthew Young
> Graduate Student
> Boston University Dept. of Astronomy
> --------------------------------------------------------------
>
>    ------------------------------
> *From:* Mark Adams [mfadams at lbl.gov]
> *Sent:* Wednesday, June 10, 2015 10:33 AM
> *To:* Young, Matthew, Adam
> *Subject:* Re: GAMG
>
>   The manual has material on this.  Let me know if you want more detail,
> algorithmic, etc.
> Mark
>
> On Wed, Jun 10, 2015 at 10:15 AM, Young, Matthew, Adam <may at bu.edu> wrote:
>
>>  Hi Mark,
>>
>>  We recently had a brief discussion (along with Matt Knepley and Jed
>> Brown) on the petsc-users list about a problem I'm working on for my
>> thesis research. Since I'd like to explore what GAMG has to offer, I was
>> wondering if you can direct me to an introduction of sorts if such a
>> document exists.
>>
>>  Best,
>> Matt
>>
>>  --------------------------------------------------------------
>> Matthew Young
>> Graduate Student
>> Boston University Dept. of Astronomy
>> --------------------------------------------------------------
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20150610/94c4c6f8/attachment.html>


More information about the petsc-users mailing list